Environmental Sound Recognition With Time–Frequency Audio Features

Chu, Selina; Narayanan, Shrikanth; Kuo, C.-C. Jay

doi:10.1109/tasl.2009.2017438

Cited by 540 publications

(387 citation statements)

References 27 publications

Supporting

Mentioning

376

Contrasting

Unclassified

Order By: Relevance

“…The time-frequency spectrogram is viewed as the 2-D image on which we extract the proposed features (Section 2.1) at each frequency bin via summation along the time axis, i.e., by setting D f = r = (t , f ) ∀t , f = f to produce frequency-wise feature vector x(f) in (2). The temporal information is marginalized out in order to make the feature invariant to temporal shift, while the local patterns characterize the local temporal dynamics which are effective clues to classify the non-stationary sound.…”

Section: Sound Classification Schemementioning

confidence: 99%

“…While the methods for classifying speech and music have been intensively developed for decades, those for the environmental sounds are studied with keen attention in recent years [1,2,3,4]. The environmental sounds are different from the speech and music in that the acoustic signals are not stationary nor well-structured; characteristics in these types of sounds are discussed in [5].…”

Section: Introductionmentioning

confidence: 99%

“…The environmental sounds are different from the speech and music in that the acoustic signals are not stationary nor well-structured; characteristics in these types of sounds are discussed in [5]. Therefore, the conventional speech recognition methods might not be suitable for the environmental sounds [1,2,6], and much research effort has been made especially on the feature extraction that effectively characterize them.…”

Section: Introductionmentioning

confidence: 99%

“…Wavelet packet analysis is applied to extract acoustic feature in generalized sound classification [7]. Recently, the method of matching pursuit that sparsely decomposes the signal by using over-complete dictionaries has been successfully applied to classify the environmental sounds [2,5]. As in the matching pursuit, the non-negative matrix factorization (NMF) also works on sparse factorization of signals with learning the dictionary; Cotton and Ellis [8] employ the NMF to construct acoustic event-based patch features from a spectrogram.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Acoustic feature extraction by statistics based local binary pattern for environmental sound classification

Kobayashi

2014

2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

Classification of environmental sounds is a fundamental procedure for a wide range of real-world applications. In this paper, we propose a novel acoustic feature extraction method for classifying the environmental sounds. The proposed method is motivated from the image processing technique, local binary pattern (LBP), and works on a spectrogram which forms two-dimensional (time-frequency) data like an image. Since the spectrogram contains noisy pixel values, for improving classification performance, it is crucial to extract the features which are robust to the fluctuations in pixel values. We effectively incorporate the local statistics, mean and standard deviation on local pixels, to establish robust LBP. In addition, we provide the technique of L 2 -Hellinger normalization which is efficiently applied to the proposed features so as to further enhance the discriminative power while increasing the robustness. In the experiments on environmental sound classification using RWCP dataset that contains 105 sound categories, the proposed method produces the superior performance (98.62%) compared to the other methods, exhibiting significant improvements over the standard LBP method as well as robustness to noise and low computation time.

show abstract

Section: Sound Classification Schemementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Acoustic feature extraction by statistics based local binary pattern for environmental sound classification

Kobayashi

2014

2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

show abstract

“…characterization of environment by typical sounds it contains) there is research in sound parameterization of various animal species which improves speech recognition [11]. There is also research in various specific applications of knowledge of relevant objective and subjective parameters of soundscape in artificial sound synthesis for consumer electronics and robotics [12][13][14][15][16][17] or as a tool for the visually impaired [18].…”

Section: Introductionmentioning

confidence: 99%

Implementation of Bipolar Adjective Pairs in Analysis of Urban Acoustic Environments

Suhanek

Đurek

2016

PROMET

View full text Add to dashboard Cite

show abstract

Ultralow Quiescent Power‐Consumption Wake‐Up Technology Based on the Bionic Triboelectric Nanogenerator

et al. 2020

View full text Add to dashboard Cite

Wake‐up circuits in smart microsystems make huge contributions to energy conservation of electronic networks in unmanned areas, which still require higher pressure‐triggering sensitivity and lower power consumption. In this work, a bionic triboelectric nanogenerator (bTENG) is developed to serve as a self‐powered motion sensor in the wake‐up circuit, which captures slight mechanical disturbances and overcomes the drawback of conventional self‐powered motion sensors in the wake‐up circuit that the circuit can only be triggered when a considerable pressure is applied on the sensor. The bTENG mimics the structure of plants and the addition of the leaf‐shaped tentacle structures can increase the electrical outputs by four times, which largely extends the detection range of the wake‐up circuit. The bTENG can detect both noncontact and contact mechanical disturbances; and voltages generated from both situations can trigger the wake‐up system. Moreover, the specially designed circuit that is compatible with the bTENG can help more accurately control the wake‐up system and prolong the battery life of the electronic networks to 12.4 times. An intrusion detection system is established in the wake‐up circuit to distinguish human motion and judge the scene. This work opens new horizons for wake‐up technologies, and provides new routes for persistent sensing.

show abstract

Environmental Sound Recognition With Time–Frequency Audio Features

Cited by 540 publications

References 27 publications

Acoustic feature extraction by statistics based local binary pattern for environmental sound classification

Acoustic feature extraction by statistics based local binary pattern for environmental sound classification

Implementation of Bipolar Adjective Pairs in Analysis of Urban Acoustic Environments

Ultralow Quiescent Power‐Consumption Wake‐Up Technology Based on the Bionic Triboelectric Nanogenerator

Contact Info

Product

Resources

About